metropolis: MCMC/Metropolis algorithm for solving Hamiltonian functions

In CRC1266-A2/moin: Modelling Interactions

Description

Description

Usage

metropolis(hfunc, hvars, hconsts, hvar_constraints, beta = 100,
  beta_prod = 2, threshold = 0.001, threshold_window = 50)

Arguments

hfunc	a Hamiltonian function either of the Ariadne or one of the gravity models
hvars	a list of named arguments to the selected Hamiltonian function
hconsts	a list of named constants to the selected Hamiltonian function
hvar_constraints	a list with named vectors giving the allowed range for the constraining variables
beta	a scalar; the cooling factor of the "hot Hamiltonian landscape"
beta_prod	a scalar; the factor to increase beta
threshold	a scalar; stopping threshold
threshold_window	a scalar; the size of the moving average window calculating the stopping threshold

Value

A list of class moin_hamiltonian containing the following components:

hvars	list of hvars with their equilibrium values
hconsts	list of hconsts supplied to the model
parms	named vector of the model parameters
iterations	data.frame containing the values of the Hamiltonian function (H) and the moving average of H (meanH) at each iteration

Author(s)

Daniel Knitter <[e-mail]>

Joe Roe <[e-mail]>

Ray Rivers <[e-mail]>

CRC1266-A2/moin documentation built on May 7, 2019, 8:56 p.m.